v0.8.1

New Features

Compatibility with NIR

• add Synaptic support for CubaLIF, update tau/r for Leaky by @pengzhouzp in #227
• Update NIR <> snnTorch by @stevenabreu7 in #246

Compatibility with torch.compile()

• Add support for torch.compile(fullgraph=True) WIP by @gekkom in #271
• Add torch.compile(fullgraph=True) support for more neuron models by @gekkom in #292
• Remove default values in autograd functions that break torch.compile by @gekkom in #291

Leaky Parallel Neurons

• LeakyParallel Neuron Implementation by @jeshraghian in #266
• Parallel by @jeshraghian in #288

Custom Surrogate Gradients

• Custom Surrogate Gradient Function by @mehranfaraji in #237
• remove duplicated arctangent formula by @abrasumente233 in #238
• Update snntorch.surrogate.rst by @ahenkes1 in #239

State Quantization

• State Quantization more robust for 2-bits by @SreyesVenkatesh in #231

Other Additions

• Btnn tests by @saiftyfirst in #243
• Add MPS as a possible computing backend by @laurentperrinet in #247
• Saiftyfirst/imbalanced class weighting by @saiftyfirst in #255

Bug Fixes

• Update build.yml by @jeshraghian in #233
• modify on synaptic neuron by @be-Berserker in #235

Docs + Tutorials

• Update README.rst by @jeshraghian in #241
• zh-cn Tutorial_1 Merge Request by @ruhai-lin in #245
• citation update by @jeshraghian in #249
• Update tutorial_5_FCN.ipynb by @jeshraghian in #251
• Update README.rst by @jeshraghian in #259
• Add DVS Gesture Tutorial by @DerrickL25 in #264
• Update tutorial_regression_1.rst by @seanvenadas in #268
• Update tutorial_regression_1.rst by @seanvenadas in #269
• BSNN tutorial by @mercadoerik1031 in #272
• Exoplanet Hunter & zh-cn tutorials Uploaded by @ruhai-lin in #281
• using tonic in place of spikedata by @laurentperrinet in #267
• adding George's POKERDVS Tutorial by @george203 in #295
• STMNIST addition to snntorch by @shatoparbabanerjee in #273
• Optimized and meticulously refined tutorial materials, meticulously crafted for comprehension and effectiveness. by @djlouie in #297

New Contributors

• @be-Berserker made their first contribution in #235
• @mehranfaraji made their first contribution in #237
• @abrasumente233 made their first contribution in #238
• @ruhai-lin made their first contribution in #245
• @laurentperrinet made their first contribution in #247
• @saiftyfirst made their first contribution in #243
• @DerrickL25 made their first contribution in #264
• @seanvenadas made their first contribution in #268
• @mercadoerik1031 made their first contribution in #272
• @gekkom made their first contribution in #271
• @stevenabreu7 made their first contribution in #246
• @george203 made their first contribution in #295
• @shatoparbabanerjee made their first contribution in #273
• @djlouie made their first contribution in #297

Full Changelog: v0.7.0...v0.8.0

v0.7.0

The biggest addition is the snntorch.export module that interfaces snnTorch modules with the Neuromorphic Intermediate Representation. SNN models trained in various libraries (e.g., Norse, Rockpool, Sinabs, lava-dl, etc.) can be converted in order to take advantage of the backends available in specific libraries.

What's Changed

• Update leaky.py by @ahenkes1 in #217
• Graded spikes for Leaky by @ahenkes1 in #218
• Fix graded_spikes registration by @ahenkes1 in #220
• feat(neuron, synaptic): add missing args in example by @gianfa in #222
• Fix regression tutorial 2 MNIST data path for linux by @timoklein in #224
• add docstrings to NIR export function by @jeshraghian in #226

New Contributors

• @timoklein made their first contribution in #224

Full Changelog: v0.6.4...v0.7.0

v0.6.0

What's Changed

• Recurrent neuron models have been added and revamped
• backprop.py has been deprecated
• Neuron model has been modified such that the rest function is scaled by beta. Seems to enable better performance.
• Reset mechanisms have been fixed
• spikegen and loss functions updated for macbook usage (metal performance shaders "mps")
• neuron computations fixed for DDP (thanks @ridgerchu !)
• graded spikes have been added (thanks @ahenkes1 !)
• BatchNormThroughTime (thanks @vinniesun !)
• Updated Surrogate functions & default surrogate converted to ATan (@ridgerchu again)
• Input & Output Functions (@ridgerchu again again)

New Contributors

• @ahenkes1 made their first contribution in #127
• @katywarr made their first contribution in #130
• @gianfa made their first contribution in #132
• @TrellixVulnTeam made their first contribution in #135
• @aloe8475 made their first contribution in #174

Full Changelog: v0.5.3...v0.6.0

v0.5.2

What's Changed

• leaky and rleaky state function substract function fix by @pengzhouzp in #95
• Detach and Reset Spikes in RLeaky by @manuelbre in #108
• Integrate ATan Surrogate function. by @ridgerchu in #111
• bptt bug may trigger device inconsistency. by @ridgerchu in #115
• Add a new feature 'probe' by @ridgerchu in #117

New Contributors

• @pengzhouzp made their first contribution in #95
• @manuelbre made their first contribution in #108
• @ridgerchu made their first contribution in #111
• @MegaYEye made their first contribution in #118

Full Changelog: v0.5.1...v0.5.2

alpha-4 release

What's new?

• refactored structure of neuron models to make it easier to integrate custom neurons
• added recurrent Leaky neuron RLeaky
• added recurrent Synaptic neuron RSynaptic
• Spiking LSTM neurons added SLSTM
• Spiking Convolutional 2d LSTMs added SConv2dLSTM
• learnable thresholds for all neurons
• learnable explicit recurrence
• Reset mechanism now includes 'none' as an option
• update unit tests

snntorch.surrogate

• Triangular surrogate
• Straight through estimator

snntorch.functional

• mse_temporal_loss function added
  Applies mean square error the first F spikes. Option for tolerance included, as well as passing labels to be converted into spike-time targets.

• ce_temporal_loss added
  Applies cross entropy loss to an inversion of the first spike. Inversion options include -1 * x and 1/x which means maximizing the logit of the correct class corresponds to minimizing the correct neuron's firing time.

• accuracy_temporal added
  Measures accuracy based on the occurrence of the first spike

Full Changelog: v0.4.11...v0.5.0

alpha-3 release

Some of the bugs from the previous versions have now been fixed w.r.t. sizes of tensors in spike encoding.

What's new?

snntorch.spikegen

• Data & target conversion have been separated out
• Conversion sizes have been fixed
• Time dimension is only created if tensor is time-varying (i.e., latency will always have time-dimension; rate might not)
• Latency & rate target conversion
• interpolation, on/off spike vals, time to first spike, on/off rate options included

snntorch.surrogate

• Parameterization of surrogate gradients has been removed from global variable to local variables within closures
• Spike operator (1/u)
• Leaky Local spike operator (leaky relu shifted equivalent)
• Local stochastic spike operator

alpha-2 release

Some of the bugs from the previous versions have now been fixed.

What's new?

snntorch

• SRM0 neuron model fix
• Reset now applies the threshold rather than '1'
• Reset by subtraction and reset to zero methods applied to both Stein and SRM0 neurons

snntorch.spikegen

• Delta modulation

snntorch.surrogate

• Optimized grad calculation

dev notes

• Travis-CI is no longer free. Replaced travis.yml with GH actions integration + tox

alpha-1 release

The first functional iteration of snnTorch!

What's new?

snntorch
The workhorse of the package.
All neuron models are integrated here, and a default Heaviside gradient is used to override the non-differentiability with conventional autograd methods in PyTorch.

• Stein's neuron model
• SRM0 neuron model
• firing inhibition, thanks to @xxwang1
• hidden states can optionally be initialized as instance variables if the user wants to just use a built-in backprop method

snntorch.backprop

• Backprop through time (BPTT)
• Truncated backprop through time (TBPTT)
• Real-time recurrent learning (RTRL)

snntorch.spikegen

• Poisson spike train generator
• Rate coding
• Latency coding

snntorch.surrogate

• FastSigmoid
• Sigmoid
• Spike Rate Escape

snntorch.spikeplot

• Raster plots
• Feature map animator
• Spike count animator

snntorch.utils

• Data split
• Data reduction

Plans for alpha-2

• delta & delta-sigma spike generators for snntorch.spikegen
• Simplified Stein's model (reduce hidden states from 2 to 1)
• More surrogate and backprop methods
• add more tests